Motion photos taken too late

motioneye-project / motioneyeos

A Video Surveillance OS For Single-board Computers

Other

7.85k stars 897 forks source link

Motion photos taken too late #1357

Open whybecausewhy opened 6 years ago

whybecausewhy commented 6 years ago

I have a pi zero set up with the zeroview and the camera is mounted on a window pointed toward my driveway. All of the photos that are sent to my email are either too late or way to early. I do not get a car pulling out of the garage or a person when someone walks by the window....just a nice photo of my driveway. Is there any documentation on how to set up the motion capture to get photos with the motion shown on them?

mb39924 commented 6 years ago

I have these settings which work great.

For the notification I set Attached Pictures Time Span to 4

Felixrising commented 6 years ago

What is your video device frame rate set to? That could have a big impact on the "minimum motion frames" setting and is an important part of the motion detection trigger.

whybecausewhy commented 6 years ago

Thanks mb39924. I made the changes and I will see what happens tomorrow.

Hooch76 commented 6 years ago

I do not get a car pulling out of the garage or a person when someone walks by the window....just a nice photo of my driveway.

Add me to the list, too. The same is happening for me, too. It would really be helpful if there was a description that would explain all that options. I was not able to find one, e.g.

Motion detection: Captured before / Captured after: is this only applicable for creating the movie or does it also apply to pictures?
Still images: Capture mode: Motion triggered (One Picture), Motion triggered, All frames. What's the exact difference?

Frankthedoc commented 6 years ago

Not sure if this applies to you, but I'll post as it may answer for some people: I had this problem when I was using an older Pi (B+) with a 1080p camera. Delay in capturing an image could be several minutes! My resolution was to buy an ODroid XU4, which can handle all the data coming its way (motion does a lot of CPU work). An alternative might be to use a lower resolution output from the camera (if available). Masking or dropping the resolution in motioneyeos itself didn't yield good results for me.